An AI assistant for your Windows screenshots.
Auto-generate titles, descriptions, and structured page intelligence for every step. Opt-in. Per-call USD cost tracking. No surprises.
Click the “Add member” button in the top-right of the team settings panel. The role selector defaults to Viewer.
Two modes. You pick which one runs.
Some screenshots are list views — pricing tiers, table rows, search results. Some are dense pages where prose is the right answer. List mode runs by default; switch to summary when you want a paragraph instead.
- List mode: extracts items via DOM + cached CSS selectors per URL pattern
- Summary mode: vision model returns a 1–2 paragraph caption
- Selectors are reused across the same site — Zillow listings, GitHub PRs, etc.
- Falls back to a prose bullet list if selectors fail
The pricing page shows three tiers in a side-by-side comparison. The Team plan is highlighted as the recommended choice with a blue ring and a “Most popular” tag.
Nothing runs until you say so. Every call shows the cost.
A first-run modal asks before any AI call goes out. UIHike records the USD spend of each enrichment, per step. No background processing, no quietly burned credits.
The original screenshot is never modified — AI-enriched data lives next to it, not on top of it.
Scoped to walkthroughs. Not a chat window.
This isn't a general-purpose assistant. It does four things — and does them well — for the screenshots and steps in your walkthrough.
Auto-titled steps
Picks a title from the page heading, document title, or vision model fallback.
Step descriptions
1–2 paragraph captions for dense pages, or extracted lists for tabular ones.
Page intelligence
Open Graph, headings, meta — surfaced as structured fields you can reuse.
Tags & entities
Auto-extracted classifiers for filtering and search across walkthroughs.
Stop writing “step 4” over and over.
Let UIHike draft the titles and descriptions. You edit what matters.